University of Rochester, Materials Science Program
Instrument strings are designed to vibrate at a certain pitch, and a musician can use their fingers to shorten the effective string length, changing the pitch. However, strings can start to unravel and potentially break. They are held under tension and therefore can release enough force, when broken, to break skin. Instrumentalists must be aware of these hazards to prevent a broken string snapping and potentially injuring them.
Strings typically have a core and wrappings around the core, which are visible by eye. An exception is the violin’s E string, or highest pitch string. This string is typically unwrapped, and composed of a steel core. Other cores include nylon, gut, and sometimes fluoropolymers. Winding materials include aluminum, steel, silver, and bronze. These materials used in the strings will slowly oxidize or corrode, due to the oxygen in the air as well as moisture. The oils on the player’s fingers also assist with this mechanism. Strings are often coated with polymers, like Teflon, or sometimes coated with gold or silver. These coatings slow down the degradation process.
Subsections of the string were be taken from the fingerboard region, where the musician uses their fingers to shorten the string, and the region near the bridge, where the bow drags across the strings in order to make them vibrate. Additionally, a section of string between the two subsections was taken to view a cross-section of each string. These string samples were examined using a scanning electron microscope (SEM) in two different modes: Secondary Electron Imaging and Backscatter Imaging. Then, Energy-Dispersive X-ray Microscopy was used on the cross-sections to identify each core and winding's composition.
2. Sample Preparation
Sample preparation included bare string sections and cross-sections.
I. Bare String Sections:
Each string had two samples taken from it; one from the area near the bridge of the instrument, and one from the fingerboard area. These sections were obtained using a razor blade to cut the string.
SEM clip holders were used to hold 8 string samples taken from the middle of the strings, between the bridge and fingerboard sections. These were then placed in a metal container, which was filled with acrylic. The acrylic was set by heating in an oven at 70 degrees Celsius. The acrylic was allowed to cool to room temperature. Then, a polishing machine polished the sample smooth, using increasingly finer grit polishing paper. Any scratches seen in the cross-section images are from polishing, not wear. The smallest grit used was 5 angstrom grit polishing paper. Once smooth, carbon paint was used to ground the areas surrounding the string cross-sections. Finally, the samples were sputter coated with gold. The cross-section samples were sputtered for 120 seconds, yielding a layer of gold about 120 angstroms, or 12 nm deep.
3. Result of Microscopy and Discussion
Different microscopic methods yielded information about different aspects of the string samples. In this project, three distinct modes of microscopic methods were used; (i) Secondary Electron Imaging, (ii) Backscattered Electron Imaging and Energy-Dispersive X-ray Microscopy. All of these methods were performed on the scanning electron microscope. Colorization was used to color code the images for clarity. Each color corresponds with a specific string, numbered 1-8.
I. Secondary Electron Imaging:
SEMs use electrons to obtain images, often called micrographs. A beam of electrons scans across the sample and the electrons interact with the sample. The electrons collected by detectors are collected to create images. These electrons can be collected in different ways to produce images. Secondary electrons occur when the electrons from the beam interact with the outer shells of the sample's electrons. This results in an inelastic collision, resulting in these low energy secondary electrons. These electrons are collected by a detector with a positive bias. This allows electrons further away from the detector to be used in creating an image. Secondary electron images provide information about the sample's surface morphology. Here, the secondary electrons are used to determine the type of wear present on the eight string samples.
String 1 has visible wear, seen in Figure 1. The section taken from the fingerboard has visible pits and scratches. These pits are likely from the oils on the musician's fingers eating away at the metal windings on the outside of the string. There are two types of scratches seen in this sample. There are scratches that appear horizontal with the string's direction, as seen in the upper left image in Figure 1. These scratches are likely from the string's surface being smoothed during production. However, there are additional scratches that do not line up with the direction of the string. These scratches are likely from the player's fingernails dragging across the string, deforming the metal's surface. There is a deeper scratch parallel to the direction of the string that can be seen in the image in the upper right corner of Figure 1. This scratch appears deeper than the ones that are likely from smoothing during production, as well as being more irregular. This one could be from the musician shifting positions, or sliding their hand further up the fingerboard to produce a higher note.
The lower two images in Figure 1 show the section of string 1 taken from the bridge area. The image on the lower left shows charging, which occurs when areas of the sample are not grounded. Charging results in areas of negative charge accumulating on the sample. This can lead to the scanning beam of electrons to be deflected toward or away from the secondary electron detector. When the beam is deflected towards the detector, these areas appear brighter. When the beam is deflected away from the detector, those areas appear darker. Both situations can be seen in the image on the lower left. The charging is likely caused by rosin on the surface of the string. Rosin is a type of resin used to allow the bows musicians use to grip the string and allow the string to vibrate easily. The areas the musicians use the bow in are adjacent to the bridge. So, string sections taken from the bridge area will likely have some rosin if they were not thoroughly cleaned before imaging. Since these all of the strings imaged were prepared as they were recieved, it is likely that many strings had rosin on them. Looking past the charging areas, the scratches on the string appear to be mostly parallel to the direction of the string. The image in the lower right corner of Figure 1 appears to simply have wear along the edges of the windings. No perpendicular scratches are observed, like the ones from the fingerboard area. Therefore, this sample has more wear along the fingerboard region than the bridge region.
The second string does not have windings, and has a somewhat smooth surface. The image in the upper right corner of Figure 2 shows some faint scratches in the direction of the string which came from smoothing during production However, most of the scratches and deformations were from wear after production. Figure 2 shows that the area near the fingerboard contains slightly more scratches than the area from the bridge. This slight difference does not indicate more wear occurring in one area than another in this string.
String 3 does not appear to have any difference in wear patterns between the sample taken from the fingerboard or near the bridge. In fact, Figure 3 does not show much wear at all from either section of the string. Instead, most of the scratches are from smoothing during the string's production. Only a few scratches perpendicular to the string's length are seen. There is a bit of wear along the edges of the windings, but it appears to be similar between the two sections. This string likely wasn't used very heavily before being taken off of the instrument. Therefore, nothing can be concluded about string wear patterns from this string alone.
String 4 is similar to string 3 in that it does not show much wear. Both the fingerboard and bridge regions, seen in Figure 4, appear to have only the scratches from smoothing. The edges of the windings appear tight and do not show signs of wear. This string is also not helpful in discovering the difference, if any, in wear patterns between the bridge and fingerboard regions of the string.
The fifth string is different, in that the surface smoothing appears in the direction perpendicular to the length of the string. It is irregular, making wear from playing difficult to identify. However, the edges of the windings have very visible charging. The gap between the windings appears larger than the other strings so far. Although, it is difficult to determine if this was by design, or through wear.
String 6 does show a difference in wear patterns between the fingerboard and bridge sections of the string. Figure 6 shows the section taken from the fingerboard has more scratches and deformations than the section taken from the bridge area. The bridge area appears to only have scratches indicative of the smoothing process during production. The fingerboard section shows scratches perpendicular to the string's length, as well as w large deformation in the surface, seen in the upper left image of Figure 6. Here, the difference between the fingerboard and bridge regions is more obvious. There is more wear in the fingerboard area.
String 7 has a large difference between the fingerboard and bridge regions of the string. Figure 7 shows the bridge region has rosin causing charging, resulting in difficulty observing wear near the bridge. However, there does not appear to be any scratches perpendicular to the length of the string in this area. The fingerboard area does contain some scratches that cannot be attributed to smoothing during production. These scratches are very superficial and are difficult to see in the lower magnification image. But, there is visibly more wear in the fingerboard region than the bridge region.
Lastly, string 8 has visible wear in both the fingerboard and bridge regions. The top images in Figure 8 show that there are small and superficial scratches in the fingerboard region. These are more obvious in the higher magnification image on the right. The bridge region shows that there are also scratches perpendicular to the length of the string. These scratches are regular and in the same direction. These scratches could be from the bow moving across the string, but the other samples did not show this type of wear. Perhaps the other strings were not oriented where the bow's contact with the string was visible. Again, the bridge area of the string shows charging from the rosin, but it does not obscure the wear on the string.
II. Backscatter Electron Imaging and Energy-Dispersive X-ray Microscopy:
Backscattered (BSD) electrons are caused by elastic collisions between the beam electrons and the sample. The energy of the backscattered electron depends on the nucleus of the atom it interacted with, allowing information about the relative atomic weights to be determined. The brighter an area shows in a backscattered image, the heavier the element that scattered it. Energy-Dispersive X-ray Microscopy (EDX) can identify the elemental composition of the sample. Electrons from the beam can be absorbed by electrons deep within the sample, emitting x-rays instead of scattering the electron. These x-rays can then be detected and analyzed to determine what element created the x-rays. In this way, the elemental composition of the strings can be determined.
The first string cross-section unwound before the acrylic set. However, its core and winding are still obvious. The honeycomb structure seen in the lower left in Figure 9 shows the carbon core of the string. The EDX spectra of the core was carbon, and not shown. The EDX of the winding was primarily aluminum, with a small amount of magnesium. The overall spectra show the entire composition of the string. However, the carbon peak is higher than it would be normally, as the acrylic shows up in the EDX spectrum as carbon. Additionally, a gold peak is seen. All of the overall spectra will show this peak, as the surface of the cross-sections were coated in gold to make them conductive, and reduce charging effects.
String 2's cross section yielded two areas with different composition. The EDX of each section is not shown, but the overall spectra is shown in Figure 10. The EDX spectra show that the core of the string is composed of iron, chromium, and nickel. This is the composition of stainless steel. The bright outer layer, seen in the BSD image in Figure 10, is composed of gold. In this case, the signal for gold overshot the background gold signal significantly. Again, the overall spectra shows a large carbon peak, due to the acrylic surrounding the string.
The third string contains multiple winding layers, seen in Figure 11. The two outermost windings partially unwound, but the structure is still obvious. EDX determined the core center to be composed of iron. The next two winding layers are then composed of copper. The outer layer is nickel. As with all the previous spectra, the carbon and gold peaks remain as a background signal from the acrylic and gold layer on the surface.
String 4 is very similar to string 3. Figure 12 shows the outer two windings partially unwound, like string 3. The core and winding compositions are identical as well; the core was iron, then the next two windings composed of copper, and the outer winding was nickel.
The fifth string is very different from the previous strings. The core of the string is composed of six circular rods surrounding a central core, as seen in Figure 13. These cores are all iron. EDX shows that the dark area between the core and the outer winding is composed of carbon and oxygen. This could be a carbon coating, or a polymer. The outer winding, which is partially unraveled, is composed of nickel.
String 6 also had a similar composition as strings 3 and 4. The core was iron, the next to windings were copper, and the outer winding was nickel. The overall EDX, seen in Figure 14, also shows silicon. This peak actually belongs to copper. The silicon and copper peaks are very close to each other in energy, and this peak was misattributed to silicon. Strings 3 and 4 show EDX spectra where this peak is attributed to copper.
String 7 has three layers. The central core, which appears dark in the BSD image in Figure 15, is carbon. This is likely a polymer. The next layer is composed of iron, chromium, and aluminum, which is an alloy similar to stainless steel, called Kanthal. Finally, the outer winding is silver. The overall spectrum does not show the chromium and aluminum peaks, since the carbon peak was so intense.
Finally, string 8 has only a core, no windings or coatings. The BSD image, seen in Figure 16, shows the same intensity throughout the entirety of the string. Multiple spots were picked to run EDX on, and they all remained the same. For that reason, the overall spectra shown is from a spot within the string. Therefore, any extra peaks from the acrylic are not present. The string's composition is iron, chromium, nickel, and magnesium. This is stainless steel, with some magnesium in it. The gold peak here is from the coating of gold to prevent charging.
BSD and EDX can tell the elemental composition of the strings. However, this data is only qualitative. It does not show the ratios of the elements, due to the heavier elements resulting in higher energy x-rays. But, these techniques did identify each string layer's composition.
The wear patterns from the fingerboard region tend to have more wear than the regions near the bridge. Rosin builds up on the regions near the bridge, appearing bright under the SEM due to charging effects. The rosin itself does not appear to effect the wear on the string, as no pits are seen in the bridge region of the strings. Any investigations into preventing or slowing down wear on instrument strings should be aimed at the fingerboard region, or the player's hands.
I would like to acknowledge and thank Brian McIntyre for his help and advice through the duration this project. I would also like to thank Sullivan Violins for providing strings for me to image. Photoshop Elements was used to colorize and make small adjustments to brightness and contrast to the images obtained from the SEM.